Traditionally, for customers running IP networks, they typically rely on routing protocols such as Border Gateway Protocol (BGP), Open Shortest Path First (OSPF), Routing Information Protocol (RIP) also known as layer 3 IP Routing Protocols, to detect a failure in the network and route around the problem. These routing protocols send systematic “hello” messages to their remote end points and when one isn't observed after a period of time, they assume there's a problem and route around it. So, these Layer 3 IP Routing Protocols are needed to re-route around failures, such as between two separate networks like Ethernet and ATM. Using these Layer 3 techniques takes about 30-60 seconds to detect and respond to a failure. This is inefficient and too slow but that is a limitation of the Layer 3 routing protocols.
Customers who rely on Frame Relay/ATM networks are used to failure notification response times of 1-2 seconds (or less). The FR/ATM network independently generates a failure notification towards customer premises equipment at the onset of failure. Customers of this new Ethernet to FR/ATM Interworking service require similar failure notification behavior from the network. Therefore, it is required to integrate automated Operations, Administration and Management traffic which can propagate failure indications end-to-end much faster, and does not require either a Layer 3 IP Routing Protocol or any type of “hello” or integration timer messages.
It is known that regular traffic can be interworked between Ethernet and ATM, e.g. using the technique described in the patent application Ser. No. 10/016,019. It is also known that there is a native OAM in the Ethernet domain, e.g. see the patent application Ser. No. 10/248,761 and Ser. No. 10/248,858. The difficulty involves implementing Ethernet OAM first, then bridging EOAM with ATM OAM.
So, besides integrating (interworking) regular traffic data between Ethernet and ATM, it is also advantageous to integrate OAM failure notification between the two separate networks. Moreover, it is also advantageous to provide failure indications by a much faster and more reliable means.